hungjunglu@[...].com (Hung Jung Lu) writes:

>  Alexander Schmolck <a.schmolck@[...].net> wrote:
>  > How do younger languages like IO make this point more forcefully than lisp,
>  > smalltalk etc -- languages which have been around for decades?
>  
>  Double checked with Lisp just now. And Lisp did not allow me to
>  redefine "if", "setq", "t", etc.

I presume you mean Common Lisp. I'm pretty sure that Scheme allows you
to do this without any restriction.

>  If there are tricks in Lisp to intercept "if", "setq", etc.I'd like
>  to know.

In Common Lisp you are not allowed to rediefine any external symbols
in the COMMON-LISP package:

  http://www.lisp.org/HyperSpec/Body/sec_11-1-2-1.html

and, in particular,

  http://www.lisp.org/HyperSpec/Body/sec_11-1-2-1-2.html

... so you define your own versions in another package. If you want to
know more about the issues surrounding this, then I suggest you ask on
c.l.lisp, as this is, after all, c.l.python.

>  Same if these limitations are specific to some particular
>  implementations of Lisp.

If by "Lisp" you mean "Common Lisp": no. If by "Lisp" you mean "the
Lisp language family" and by "implementation" you mean "dialect": yes.

>  I greatly suspect that these limitations of Lisp are general, since
>  Lisp does not naturally have prototype-based scopes like
>  onion-skins, and overriding things like "if" could more easily
>  create havoc.

Python does not expose pointers to the programmer. I greatly suspect
that this limitation is general to all programming languages, as this
could easily create havoc :-)